Author:
Sen Wang1,2, Ronald Clark2, Hongkai Wen2 and Niki Trigoni2
1. Edinburgh Centre for Robotics, Heriot-Watt University, UK
2. University of Oxford, UK
Download this paper: http://senwang.gitlab.io/DeepVO/#paper
Watch video: http://senwang.gitlab.io/DeepVO/#video
Call Us -/9953056974- Call Girls In Vikaspuri-/- Delhi NCR
DeepVO - Towards Visual Odometry with Deep Learning
1. DeepVO
Towards End-to-End Visual Odometry with Deep
Recurrent Convolutional Neural Networks
National Chung Cheng University, Taiwan
Robot Vision Laboratory
2017/11/08
Jacky Liu
2. About this work
DeepVO : Towards Visual Odometry with Deep Learning
Sen Wang1,2, Ronald Clark2, Hongkai Wen2 and Niki Trigoni2
1. Edinburgh Centre for Robotics, Heriot-Watt University, UK
2. University of Oxford, UK
Download this paper: http://senwang.gitlab.io/DeepVO/#paper
Watch video: http://senwang.gitlab.io/DeepVO/#video
2
DeepVO : Towards Visual Odometry with Deep Learning
3. Contributions
1. Proving that
Monocular VO could
be build by End-to-
End training
2. RCNN architecture
could generalized to
unseen environment
3. Complex movement
could be modeled by
RCNN
3
DeepVO : Towards Visual Odometry with Deep Learning
6. Network design
1. Traditional computer vision learn knowledge from
appearance and image context
2. Visual odometry should learn from geometry.
This is what RCNN tried to address
6
DeepVO : Towards Visual Odometry with Deep Learning
9. Preprocessing
Normalizing inputs (speed up training)
=> subtracting the mean RGB values of the
training set
Resize image to 64x
Stack two images to form a tensor
9
DeepVO : Towards Visual Odometry with Deep Learning
10. CNN
What this research mean by learning
“geometric” feature?
=> They stacking two RGB images and feed it
into CNN. Expecting the network to perform
feature extraction on the concatenation of
two consecutive monocular RGB images.
10
DeepVO : Towards Visual Odometry with Deep Learning
11. RNN
RNN is not suitable to directly learn sequential
representation from high-dimensional raw
data, such as images.
Hidden state:
ℎ 𝑘 = ℋ 𝑊𝑥ℎ 𝑥 𝑘 + 𝑊ℎℎℎ 𝑘−1 + 𝑏ℎ
Output:
𝑦 𝑘 = 𝑊ℎ𝑦ℎ 𝑘 + 𝑏 𝑦
11
DeepVO : Towards Visual Odometry with Deep Learning
𝑏: bias vector𝑊: weight matrix
𝑘: time index ℋ: activation function
Vanishing gradient
problem
12. LSTM (Long short-term memory)
12
DeepVO : Towards Visual Odometry with Deep Learning
Need depth to
learn high level
representation
22. Dynamic
This research don’t
know how to deal
with this issue
Traditional VO –
RANSAC (remove
outlier)
Get more training
data
22
DeepVO : Towards Visual Odometry with Deep Learning
23. Conclusion
23
End-to-end monocular VO based on Deep learning
Deep RCNN
No need to carefully tune the parameters of the
VO system
It is not expected as a replacement to the classic
geometry based approach